January 6, 2006
M3--THE FINALS DAYS
It's nearly curtains for M3, the broadest measure of money supply currently published by the Federal Reserve. "Currently" being the operative word here, since the Fed will cease M3 updates in March.
Dying, but not yet dead. Indeed, in the limited time left for M3 the series is showing some spirit for one final run skyward. In the latest weekly money-supply data published yesterday, M3 advanced by 8.1% for the week of December 26, 2005 v. its year-earlier level. The somewhat narrower gauge of money supply known as M2 grew by roughly half as fast, rising 4.1% over the same span. In fact, that's no surprise. The spread between the two rates of increase has been growing ever wider as 2005 unfolded (as the chart below reveals), suggesting that there may be more difference between M2 and M3 than the central bank likes to admit.
Officially, the Fed argues that M2 and M3 are virtually identical, and so M3 is a statistical redundancy of no material value, and therefore scheduled for the numerical scrapheap. As funerals go, this one promises to be as quiet as a bond conference in 1999.
As CS readers know, we've written previously on the impending demise of M3--first here, and then again here. And now we dive in for a third time, in part because the gap between the two measures shows every indication of widening, perhaps to the bitter end at M3's scheduled death in March.
As long the spread grows, we'll keep updating it, no doubt to the consternation of the Fed and others. Perhaps we're unnecessarily anxious, but when the most powerful central bank on the planet says it's discarding its broadest measure of money supply, and the only one that's growing at more than twice as fast as the officially stated rate of inflation, we're intrigued, to say the least.
What, you ask, is M3 made of? Sugar and spice, along with everything that comprises M2, plus these extras, as per the Fed's definition:
1) balances in institutional money market mutual funds
2) large-denomination time deposits in amounts of $100,000 or more
3) repurchase agreement liabilities of depository institutions, in denominations of $100,000 or more, on U.S. government and federal agency securities
4) Eurodollars held by U.S. addressees at foreign branches of U.S. banks worldwide and at all banking offices in the United Kingdom and Canada.
Does the distinction matter? The Fed doesn't think so. And the market's accepting the company line, or so one can argue in a world where a 10-year Treasury yields around 4.38%, as we write--materially less than the 8.1% climbing pace set by M3.
Are we a bit paranoid and obsessive over what some might term an arcane bit of monetary statistics? Perhaps. It's a dirty job, but someone's got to do it.
Posted by jp at January 6, 2006 10:06 AM
We come not to bury variance, but only to point out more pressing, and in our minds more relevant matters. The point being that there's the potential for mischief in central banking when the yield on a "risk-free" debt instrument is but half the pace of increase (dilution?) in the underlying currency in which the coupon payments are dispensed. In a perfect world, an 8% nominal advance in money supply would compensate with an offsetting 8% yield in coupon payments denominated in said currency. Alas, compensation in the real world is just a bit over 4%.
Perhaps the money supply's rate of increase will slow. Or Treasury yields will rise. Or, to adopt the Fed's line, maybe M3 doesn't really matter. Indeed, why does the Fed want us to focus on M2, which is growing by roughly the same rate as the 10-year yield? Variance is intriguing, but it won't pay the rent.
Posted by: JP at January 6, 2006 9:07 PM
what's the 10year yield to do with the % change in M3? u should compare variance to variance, if you wanna make a point, but in this case it's not possible.
#3 & 4 are interresting though.
Posted by: noname at January 6, 2006 8:28 PM